Using the Coude Feed System of the Kitt Peak National Observatory at its highest possible dispersion and a CCD detector with wide dynamic range we were able to examine in detail the radial and turbulent velocities of the Orion Nebula at the seeing limit. We were able, for the first time, to resolve multiple components of the spectral lines and fit gaussian profiles to them. Using four different observing runs, two circular areas 2' in diameter were observed, the first centered on $\theta\sp 1$ Orionis and the second centered on $\theta\sp 1$ Orionis. The velocity resolution of 1 km ${\rm s}\sp{-1}$ combined with the angular resolution of 2" allowed us to look in detail at line splitting as well as the location and measurement of high internal motions, about $-50$ km ${\rm s}\sp{-1}$ with respect to the bulk of ionized matter. Because over 900 different positions were measured we were able to use modern statistical techniques for the analysis of turbulence, so that the dispersion relation and the structure function could be calculated and compared with predictions of the Kolmogorov theory. The random motions showed correlation with spatial separation, as expected for turbulence; but, comparison with predictions of the classical theory for nebular turbulence showed poor agreement. If Kolmogorov theory applies, then the energy must be input at many scales.

Bands of ions have been observed at constant pitch angle by the Dynamics Explorer High and Low Altitude Plasma Instruments at auroral latitudes. The observed ion dispersion pattern shows lower energy ions toward the equatorward side of the band and higher energy ions toward the poleward side of the band. Ion bands have their highest energy flux at small pitch angles. The observed bands have been correlated with storm phase (by Dst) and substorm phase (by AE). Bands are more likely to occur during main storm phase than during recovery storm phase. Substorm correlations are statistically significant, but there is a hint that most bands occur during substorm recovery phase. Two models have the potential of producing ion signatures that are similar to the band feature. They are the time-of-flight mechanism and the convective dispersion mechanism. Under a time-of-flight mechanism, ions are dispersed along a magnetic field line with higher energy particles outrunning lower energy particles. Ions are dispersed perpendicular to the magnetic field under convective dispersion. A time-of-flight effect does not explain the band energy-latitude dependence observed in the southern night or northern day very well, whereas the convective dispersion mechanism easily accomplishes this. Simulations have been performed on ions under the influence of the time-of-flight and convective dispersion mechanisms. The time-of-flight simulation determines the distance to the ion source. The requirement on its position is that two different energy ions, which are detected at different times, must have been released from the source simultaneously. A composite of the computed source distances shows that source motion had to be unphysical in order to create the signature of a sample band. Simulating convective dispersion required knowledge of ion drifts. Using the measured drifts on a satellite pass which showed a band, ions were traced from one hemisphere to the other. A composite of arrival latitudes from an inverted "V" simulated source in the opposite hemisphere, nearly approximates the banded ion pattern. Thus, we can conclude that convective dispersion is the correct mechanism for formation of ion bands.

A thermonuclear runaway on the surface of a degenerate white dwarf has been shown to be a plausible scenario for the classical nova phenomenon. This thesis investigates the potential for observing gamma ray line emission from novae, a unique signature of such thermonuclear runaway models. In the high temperature nuclear burning, the low mass nuclei are arranged such that positron unstable nuclei become the dominant isotopes. Upon decay these nuclei produce gamma ray photons either from de-excitation of excited states or from annihilation of the emitted positrons with electrons. To evaluate the emerging gamma ray fluxes, simple models of the expansion of the nova atmospheres are evolved. These models yield the physical conditions of the outer atmospheres which serve as input to investigations into the fate of the positrons and the transfer of the photons. The latter is accomplished with Monte Carlo simulations of the photon histories. The resulting estimates suggest that nearby fast novae could yield detectable fluxes of annihilation gamma rays from positrons produced by decay of $\sp{13}$N and $\sp{18}$F. It seems less likely that detectable fluxes of de-excitation photons would escape the nova atmosphere at early times, however significant quantities of long-lived nuclei might be produced (i.e., $\sp{22}$Na and $\sp{26}$Al) and yield characteristic gamma rays from the accumulation of the radioactive debris of many novae. Thus novae appear to be a prime target for the Oriented Scintillation Spectrometer Experiment on Gamma Ray Observatory and to offer a potential for the use of the Solar Maximum Mission Gamma Ray Spectrometer as an astronomical observatory.

The control of the strength and pattern of magnetospheric convection by the interplanetary magnetic field (IMF) is well established. This control is widely interpreted as evidence for magnetic interconnection between the IMF and the magnetosphere. A quantitative model of a magnetically interconnected ("open") magnetosphere is developed in this thesis. The model is based on the perturbation of a closed magnetic field configuration with a given magnetopause geometry. The internal field is the result of a dipole field confined by Chapman-Ferraro currents at the magnetopause and stretched on the nightside to produce a tail field. The external field is taken from a model based on gas-dynamic supersonic flow around a blunt object with an embedded magnetic field. The normal component at the magnetopause must be time independent and it is derived from two different hypotheses of magnetopause merging. The interconnection field is derived as a solution to a Neumann boundary value problem consisting of Laplace's equation with the magnetopause normal component distribution as a boundary condition. This interconnection field is then added to the closed models to produce the desired open configuration. The resulting open magnetospheric configuration is used to map magnetopause electric fields down to the ionosphere in the polar cap thus producing electrostatic potential contours. Various distributions result as a function of the IMF orientation and of merging hypothesis. The results can be compared with existing data to test and refine the model.

We present the first quantitative analysis of the morphology and gaseous-phase properties of two wind-blown shells and their interaction with the interstellar medium by using high spatial resolution emission-line imagery and optical spectrophotometry. We describe the spatially resolved ionization structure and analyze the physical mechanisms responsible for the observed emission by a detailed comparison of emission characteristics with those predicted by both shock-wave and photoionization models. In NGC 7635, our imagery reveals variations in density, temperature, and ionization structure on scales ranging from the size of the nebula (6' x 12') down to the seeing limit of approximately 2.5" (0.03 pc at a distance of 2.6 kpc). The overall emission in the nebula is dominated by the overwhelming contribution of the HII region and is characteristic of photoionization processes. However, when the HII region contribution is removed, the emission in the bubble region is consistent with that produced in the incomplete cooling region behind a stellar-wind shock wave. The embedded, photoevaporating cloud in NGC 7635 contributes enough mass over a dynamical lifetime to account for the shell mass of 5.0 M$\sb{\odot}$. We interpret the "Bubble Nebula" as a blister HII region containing a young stellar-wind bubble formed around a post-main-sequence O star. In NGC 2359, our imagery reveals variations in density, temperature, and ionization structure on scales ranging from the size of the nebula (14' x 14') down to the seeing limit of approximately 2.1" (0.05 pc at a distance of 5.0 kpc). The structure of the HII region can be understood in terms of a photoionized conical cavity protruding into the surrounding molecular cloud. The emission in the bubble region is characteristic of that produced in the incomplete cooling region behind a stellar-wind shock wave. The results of our spectrophotometry reveal an (OIII) temperature of approximately 27,000K for the western rim of the bubble, indicative of shock-excited emission. We interpret NGC 2359 as a blister HII region containing a stellar-wind bubble formed at the edge of a dense, molecular cloud.

Motion in the O$\sp+$ zone of the core of the Orion Nebula has been studied utilizing new high resolution CCD observations of the (O II) $\lambda\lambda$3726,3729 doublet. Emission associated with the main body of ionized gas inside the HII region consists of a narrow component due to direct emission and a broad reflected component globally redshifted relative to the narrow component. Line splitting blueshifted relative to the main emission has been found to the SE over an area more extensive than previously observed. We have derived the electron density from the (O II) ratio. A radial gradient peaked 45" SW of $\Theta\sp1\rm C$ Ori has been found for the main body of ionized gas. Relative to the main component the density of the split component appears to be considerably lower. The effective depth of the main emission line region is shown to be comparable to the distance between the dominate ionizing star $\Theta\sp1\rm C$ Ori and the principal ionization front. This component exhibits a radial gradient in depth centered to the SW. A rough estimate for the region emitting the split component suggests that its effective depth is comparable to that of the main region. Significant turbulent broadening exists in all components. Statistical analysis shows limited correlation over small scale sizes $\sim$10" not in accordance with the predictions of Kolmogorov theory. This discrepancy may be due to the observed curvature in the main ionization front which has not been considered in simple plane parallel models for HII region turbulence. Refinements to a recent model of Orion suggested by our data are proposed. A foreground "hinged" lid structure redirects the flow inside the nebula. The split component arises in material flowing away from an ionization front eroding the near side of the lid. Evidence that the background ionization front has an S-shaped geometry is presented. We argue that the ionization structure of the core is related to the proper motion of $\Theta\sp1\rm C$ to the SW relative to the molecular cloud.

As part of a large-scale search for coherent pulsations from LMXBs in the RXTE archive, I have completed a detailed series of searches for coherent pulsations of 4U 1820--30---an ultracompact LMXB with a binary period of 11.4 minutes located in the globular cluster NGC 6624. The small binary period leads to a very high acceleration, so I used phase modulation searches as well as acceleration searches to give significant sensitivity to millisecond pulsations. I searched a total of 34 archival RXTE observations, 32 of which had an on-source integration time longer than 10 ks, and some of which were made consecutively which allowed me to combine them. While I found no pulsations, I have been able to place the first stringent (95% confidence) pulsed fraction limits of ≲ 0.8% for all realistic spin frequencies (i.e. ≲ 1 kHz) and likely companion masses (0.02 M⊙ ≤ Mc≤ 0.3 M⊙ ). By contrast all five LMXBs known to emit coherent pulsations have intrinsic pulsed fractions in the range 3% to 7% when pulsations are observed.

The Solar Tower Atmospheric Cherenkov Effect Experiment (STACEE) is a ground based gamma ray telescope which studies active galatic nuclei (AGNs). As a ground based experiment, data quality is directly related to atmospheric conditions. This thesis describes the process of verifying the reliability of astronomical weather predictions made by the Canadian Meteorological Service (CMS). This includes a description of the construction, calibration, and operation of a cloud detector at McGill University in Montreal, Canada. Cloud cover data was collected from the beginning of September, 2003, to the beginning of November 2003. These data were compared to cloud cover weather predictions made by the CMS. A blind analysis consisted of scaling the CMS and real data, and calculating the correlation coefficient between these two sets of data. The correlation between the predictions and real sky conditions was 0.4622, well above the 0.1% significance level of 0.1906 Thus the CMS is determined to be a reliable weather predictor, and useful to the STACEE group.

A method for obtaining a smaller bound for the post-Newtonian parameter alpha 2 is outlined. alpha2 is a parameter in the post-Newtonian formalism associated with violation of Lorentz invariance. The current bound found by Nordvedt, alpha2 < 1.2 x 10 -7, was obtained using solar system data. We show that, by analysing millisecond pulsar profiles and limiting the amount of secular evolution, a smaller bound can be obtained.

We report on the long-term phase-coherent timing of two young, energetic pulsars, PSR B1509-58 and PSR B0540-69, using data from the Rossi X-ray Timing Explorer, the Parkes Radio Telescope and the Molonglo Observatory Synthesis Telescope. We present fully phase-coherent timing analyses of both sources, and show that contamination from timing noise can significantly affect measured timing parameters. As a result, we also present, a partially phase-coherent timing analysis which is less sensitive to timing noise. / We present an updated measurement of n = 2.839 +/- 0.003 for PSR B1509-58, and we report a measurement of the second braking index, m = 18.3 +/- 2.9, in agreement with the prediction from the simple model of pulsar spin-down. We also examine changes in dispersion measure and present an analysis of the noise, measuring a low-resolution power spectrum corresponding to a spectral index of beta = -4.6 +/- 1.0. This implies that the noise cannot be explained by a pure random walk. We also show that no glitches have occurred in 21.3 yr, implying that this pulsar glitches less often than typical young pulsars. / We measure the braking index for PSR B0540-69, n = 2.140 +/- 0.009 and discuss our measurement in the context of other discordant values reported in the literature. In addition, we detect evidence for a glitch which has been previously reported but later disputed. The glitch occurred at MJD 51335 +/- 12 with Deltanu/nu = (1.4 +/- 0.2) x 10-9 and Delta n&d2;/n&d2; = (1.33 +/- 0.02) x 10-4. We present an improved source position from the phase-coherent timing of this pulsar, to our knowledge, the first of its kind from X-ray pulsar timing. / We discuss the physical implications of n < 3, the value predicted from the model of pulsar spin down. In addition, we discuss possible causes of timing noise in these two young pulsars. We also discuss the anomalously small glitch activity in very young pulsars which suggests that neutron stars of similar apparent ages, rotation properties and inferred dipolar B fields can have significantly different internal properties.